On Mon, Feb 16, 2026 at 8:21 AM Amir Goldstein <[email protected]> wrote: > > On Thu, Feb 12, 2026 at 01:58:11PM -0800, T.J. Mercier wrote: > > This series adds support for IN_DELETE_SELF and IN_IGNORED inotify > > events to kernfs files. > > > > Currently, kernfs (used by cgroup and others) supports IN_MODIFY events > > but fails to notify watchers when the file is removed (e.g. during > > cgroup destruction). This forces userspace monitors to maintain resource > > intensive side-channels like pidfds, procfs polling, or redundant > > directory watches to detect when a cgroup dies and a watched file is > > removed. > > > > By generating IN_DELETE_SELF events on destruction, we allow watchers to > > rely on a single watch descriptor for the entire lifecycle of the > > monitored file, reducing resource usage (file descriptors, CPU cycles) > > and complexity in userspace. > > > > The series is structured as follows: > > Patch 1 refactors kernfs_elem_attr to support arbitrary event types. > > Patch 2 implements the logic to generate DELETE_SELF and IGNORED events > > on file removal. > > Patch 3 adds selftests to verify the new behavior. > > > > --- > > Changes in v2: > > Remove unused variables from new selftests per kernel test robot > > Fix kernfs_type argument per Tejun > > Inline checks for FS_MODIFY, FS_DELETE in kernfs_notify_workfn per Tejun > > > > T.J. Mercier (3): > > kernfs: allow passing fsnotify event types > > kernfs: send IN_DELETE_SELF and IN_IGNORED on file deletion > > selftests: memcg: Add tests IN_DELETE_SELF and IN_IGNORED on > > memory.events > > > > fs/kernfs/dir.c | 21 +++ > > fs/kernfs/file.c | 20 ++- > > fs/kernfs/kernfs-internal.h | 3 + > > include/linux/kernfs.h | 1 + > > .../selftests/cgroup/test_memcontrol.c | 122 ++++++++++++++++++ > > 5 files changed, 161 insertions(+), 6 deletions(-) > > > > > > base-commit: ba268514ea14b44570030e8ed2aef92a38679e85 > > -- > > 2.53.0.273.g2a3d683680-goog > > > > In future posts, please CC inotify patches to fsdevel and inotify maintainers.
Got it, will do.
